Summary Sepsis is a major health issue with a mortality rate of 30%. During periods of septic stress, glucocorticoid (GC) production is markedly induced in relation to an increased demand. However, the function of this inducible GC (iGC) remains poorly understood. Importantly, 25-60% of septic patients suffer relative adrenal insufficiency ? insufficient iGC production in response to stress. Given the potential complications of adrenal insufficiency, GC therapy is often used for septic shock patients. However, the true contribution of adrenal insufficiency to sepsis and the efficacy of GC therapy are highly controversial. While a number of factors contribute to the debate, the lack of a relative adrenal insufficiency animal model has limited our capacity to address these issues. Toward a solution, scavenger receptor BI (SR-BI), a well-established high density lipoprotein (HDL) receptor, was first identified in our laboratory as a critical protective factor in sepsis. A number of laboratories including ours recently reported that SR-BI null mice fail to produce iGC in response to ACTH stimulation, establishing SR-BI null mice as a relative adrenal insufficiency model. Using this unique model and cecal ligation and puncture (CLP)-induced sepsis, we demonstrated that mice with relative adrenal insufficiency are susceptible to CLP-induced septic death, and more importantly, supplementation of GC rescued mice with relative adrenal insufficiency, but surprisingly, caused more death in mice without relative adrenal insufficiency. These findings provide a ?proof of concept? that GC therapy may selectively benefits septic patients with relative adrenal insufficiency. In this application, we propose to use our newly developed adrenal-specific SR-BI null mice as a unique relative adrenal insufficiency animal model to elucidate the mechanism of relative adrenal insufficiency as a risk factor for sepsis and to provide mechanistic support for the selective use of GC therapy in a subgroup of patients with relative adrenal insufficiency. The translation of this preclinical study will improve the overall efficacy of sepsis therapy.